DESPITE (SPITFIRE) simulating vegetation impact and ... DESPITE (SPITFIRE) simulating vegetation

  • View

  • Download

Embed Size (px)

Text of DESPITE (SPITFIRE) simulating vegetation impact and ... DESPITE (SPITFIRE) simulating vegetation

  • DESPITE (SPITFIRE) simulating vegetation impact and emissions of wildfires

    Veiko Lehsten

    Martin Sykes

    Almut Arneth

    ESF Exploratory Workshop Farnham

  • LpJ Intro

    LPJ-GUESS (Lund Potsdam Jena General Ecosystem Simulator)

    Dynamic vegetation model

    Simulates vegetation depending on:

    Precipitation Temperature

    Radiation Soil

    Vegetation simulated in age cohorts of plant functional types e.g. tropical broad-leafed raingreen tree or C4 grass of patches of ca. 1000m2

    Stochastic effects (disturbance, senescence) covered by repeated simulation of single patches and averaging

    Smith B et al. (2001) Global Ecology and Biogeography 10, 621-637.

    ESF Exploratory Workshop Farnham

  • Litter (amount and quality)

    Crown height and length

    Stem diameter

    Plant types

    Mortality and damage per plant type and age cohort Wind

    CO2 CO NOx Methan


    Climate historic or projected



    (replicated stands)

    CO2 DESPITE Burned area: prescribed from RS or dynamic projected

    SPITFIRE developed by Kirsten Thonicke and Allan Spessa

  • LPJ-GUESS is a point model, there is no spatial heterogeneity within a cell.

    50% burned area

    To account for stochastic processes several replicates of each cell are simulated.

    Burning of simulated cells in LPJ-GUESS-DESPITE

    ESF Exploratory Workshop Farnham

    Both cells burn 50%, results in more resources for all surviving trees.

    Very limited effect on vegetation

    One cell burns 100%, the other does not burn surviving trees at unburned replicates have not more resources available.

    Strong effect on vegetation of the burned patches

  • Emissions: Results annual averages for Africa using L3JRC burned area (RS data)

    Our estimate van der Werf (2005)

    NHA 274±37 627±75 [Tg C a-1] SHA 448±75 576±72 [Tg C a-1]

    High burned area ratios lead to low accumulation of fuel

    ESF Exploratory Workshop Farnham

  • Fire effects: Emissions (Fire flux):

    Fire flux =f(fuel load, fire speed, FBD, wind, fuel moisture, fuel composition)

    LPJ-GUESS (Globfirm) Emissions=emission factor(PFT, trace gas)*Biomass burned (PFT)

    Emission of trace gases depend on fire intensity

    different in smoldering vs. burning

    Additionally to the standard emission factors, the ‘mixed fuel’ is incorporated.

    grasscombustedlittercombusted grasscombusted

    __ _

    Emission factor = f

    ESF Exploratory Workshop Farnham

  • Fire Fluxes: The mixed fuel model1 Emission Factor =f(mix of combusted fuel)


    *11.085.0  

      

     

     grasslitter


    CE : fraction of carbon emitted as CO2 relative to total carbon emission

    grass : combusted grass litter [dry weight] litter + grass: total combustion [dry weight]

    CEbaE sss *

    ES : specific emission factor aS;bS: trace gas spec. parameter

    0 0.2 0.4 0.6 0.8 1 0.84







    0 0.2 0.4 0.6 0.8 1 1







    grasslitter grass 

    C o m

    b u st

    io n e

    ff ic

    ie n cy

    [F ra

    ct io

    n o

    f ca

    rb o n e

    m m

    it te

    d as

    C O

    2 ]

    grasslitter grass 

    Combustion Efficiency (CE)

    Emission factor for CO and CH4

    0 40








    CO CH4 [g/kg]

    1Scholes, R. J., et al. (1996), Emissions of trace gases and aerosol particles due to vegetation burning in southern hemisphere Africa, JGR-Atmospheres, 101, 23677-23682.

  • Simulation with 1000 years spinup, 100 patches

    Grass FPC Tree FPC

    Fire return interval

    Fire return intervalLatitude / MAP [°/mm a-1] Latitude / MAP [°/mm a-1]

    0 5 10 15 20 25 0




    2000 mean annual precipitation

    m m

    a -1

    degree north

    Range used

    Artificial imposed fire frequency along a precipitation gradient

    ESF Exploratory Workshop Farnham

  • Better understanding of mechanisms can lead to increased performance in the simulation of vegetation types

    Simulation of vegetation at continental scale

    Fire: Prescribing MODIS burned area MCD45 transformed to fire frequencies per 1 degree grid cell

    Rooting: LPJ Standard

    Burn Frequency Average 2000-2007 MCD45

    ESF Exploratory Workshop Farnham

  • 0.23

    Continental vegetation simulation: W

    ith fi

    re (M

    O D

    IS )

    W ith

    ou t f


    Potential vegetation according to Ramankutty and Foley (1999)

    R am

    an ku

    tt y

    et a

    l. (

    1 9 9 9 )

    G lo

    b al

    B io

    g eo

    ch em

    ic al

    C yc

    le s 1 3

    , 9 9 7 -1

    0 2 7 .



    0.110.12Open Shrubland

    0.23Dense Shrubland

    0.150.15Grassland / Steppe

    0.710.46Savanna / tr deciduous forest

    0.670.49Tropical rain forest

    Biome Kappa Kappa

    Fire frequency Prescribed from Remote sensing MODIS MCD45

    ESF Exploratory Workshop Farnham

  • FUME

    Forest fires under climate, social and economic changes in Europe,

    the Mediterranean and other fire- affected areas of the world

    A Arneth, Lund, Physical Geography & Ecosystem Analysis

    ESF Exploratory Workshop Farnham

  • • 33 project partners (incl. USA, Australia, South Africa) • Coordinator: Universidad Castilla-La Mancha (José Moreno) • Status: in negotiation, expected to commence in 2010 • FP7, Large-scale integrating project • Chief goals & objectives:

    A Arneth, Lund, Physical Geography & Ecosystem Analysis

    (i) Analyse past interactions of climate, socioeconomy, LULC change, on fire regimes

    (ii) Develop new climate/socioeconomic scenarios for future projections of fire regimes (fire danger, vegetation type, fuel load, intensity, seasonality…); identify possible new fire prone areas

    (iii) Adaptation potential, fire management – strategies for fire prevention and planning, including cost assessment and new policies


    ESF Exploratory Workshop Farnham

  • A Arneth, Lund, Physical Geography & Ecosystem Analysis

    FUME 3 modules, reflecting these chief objectives:

    Recent trends in landscape change and fire occurrence: disentangling the role of climate and other factors on fire (Mod. 1)

    Projections of future fire risk and fire regime due to climate and other social and economic changes (Mod. 2)

    Adapting to change: new approaches and procedures to manage risks and landscapes under climate and social change to reduce vulnerability to fire (Mod. 3)

    ESF Exploratory Workshop Farnham

  • FUME • Main target region: Mediterranean Europe, from spot to landscape level; rural/urban interface • Europe as a whole with boreal and central Europe as expected new fire-prone regions • Include specifically climate extremes

    A Arneth, Lund, Physical Geography & Ecosystem Analysis

    ULUND: coupled detailed vegetation model (LPJ-GUESS) + Spitfire to reproduce past and assess future fire regimes, specifically vegetation-climate-fire interactions

    Spin-off (note: fire-chemistry interactions are not within the scope of the IP): fire emissions

    ESF Exploratory Workshop Farnham

  • Aims of DESPITE application: vegetation and emission simulation for fire scenarios for Europe

    with focus areas: Peak District,

    Mediterranean areas and boreal forests

    Final aim is the application of simulation results in a decision making tool for politicians to inform about the expected effects of climate change and fire management, e.g. fires to reduce fire risk or fire

    onset at different times of the year

    ESF Exploratory Workshop Farnham

  • DESPITE (SPITFIRE) simulating vegetation impact and emissions of wildfires

    Benefits from future research:

    Veiko Lehsten

    Martin Sykes

    Almut Arneth

    ESF Exploratory Workshop Farnham

    Validation of the simulations of recent: emissions combustion completeness and efficiency

    Validation of recent emissions is required for scenario simulations

    Improved emission factors related to burning conditions: Fuel type (PFT & Size) Fuel moisture Wind